At the Intersection of Health, Health Care and Policy
Cite this article as:
Joy L. Lee, Matthew Maciejewski, Shveta Raju, William H. Shrank and Niteesh K.
Choudhry
Value-Based Insurance Design: Quality Improvement But No Cost Savings
Health Affairs, 32, no.7 (2013):1251-1257
doi: 10.1377/hlthaff.2012.0902
The online version of this article, along with updated information and services, is
available at:
http://content.healthaffairs.org/content/32/7/1251.full.html
For Reprints, Links & Permissions:
http://healthaffairs.org/1340_reprints.php
E-mail Alerts : http://content.healthaffairs.org/subscriptions/etoc.dtl
To Subscribe: http://content.healthaffairs.org/subscriptions/online.shtml
Health Affairs is published monthly by Project HOPE at 7500 Old Georgetown Road, Suite 600,
Bethesda, MD 20814-6133. Copyright © 2013 by Project HOPE - The People-to-People Health
Foundation. As provided by United States copyright law (Title 17, U.S. Code), no part of Health
Affairs may be reproduced, displayed, or transmitted in any form or by any means, electronic or
mechanical, including photocopying or by information storage or retrieval systems, without prior
written permission from the Publisher. All rights reserved.
Not for commercial use or unauthorized distribution
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
Payment Models
By Joy L. Lee, Matthew Maciejewski, Shveta Raju, William H. Shrank, and Niteesh K. Choudhry
10.1377/hlthaff.2012.0902
HEALTH AFFAIRS 32,
NO. 7 (2013): 1251–1257
©2013 Project HOPE—
The People-to-People Health
Foundation, Inc.
doi:
Value-Based Insurance Design:
Quality Improvement But No
Cost Savings
Joy L. Lee is a doctoral
student at the Johns Hopkins
Bloomberg School of Public
Health, in Baltimore,
Maryland, and a research
trainee in the Division of
Pharmacoepidemiology and
Pharmacoeconomics, Brigham
and Women’s Hospital, in
Boston, Massachusetts.
Value-based insurance design (VBID) is an approach that
attempts to improve the quality of care by selectively encouraging or
discouraging the use of specific health care services, based on their
potential benefit to patients’ health, relative to their cost. Lowering
beneficiary cost sharing or out-of-pocket spending to increase medication
adherence is one common element of value-based insurance design. We
conducted a systematic review of the peer-reviewed literature to evaluate
the evidence of the effects of VBID policies on medication adherence and
medical expenditures. We identified thirteen studies assessing the effects
of VBID programs and found that the programs were consistently
associated with improved adherence (average change of 3.0 percent over
one year), as well as with lower out-of-pocket spending for drugs. In the
studies we reviewed, providing more generous coverage did not lead to
significant changes in overall medical spending for patients and insurers.
Further research is needed to understand how best to structure VBID
programs to both improve quality and reduce spending.
ABSTRACT
T
he underuse of effective drug therapies is a major contributor to suboptimal disease control and poor
outcomes for patients with chronic
conditions. Although underprescribing by providers is common,1 patients’
long-term nonadherence to prescribed medications appears particularly problematic.2 Valuebased insurance design (VBID), or evidencebased plan design, attempts to promote
adherence by lowering patients’ out-of-pocket
spending for treatments that are associated
with large reductions in illness, death, or both.
This tool is distinct from traditional insurance
design, which prices medications based only on
their cost, so that patients face low cost sharing
for low-cost medications, regardless of their efficacy. In value-based insurance design, high value
may be determined based upon both the treatment itself and the patients who are being targeted. So, for example, statin copayments may be
Matthew Maciejewski is a
professor at the Center for
Health Services Research in
Primary Care, Department of
Internal Medicine, Duke
University School of Medicine,
in Durham, North Carolina.
Shveta Raju is an internal
medicine physician at the
Center for Health Services
Research in Primary Care.
lowered for patients with coronary artery disease
but not for those receiving treatment for primary
prevention.
Employers have been experimenting with
VBID policies for a decade,3,4 as a previous review
by Niteesh Choudhry and colleagues described.5
The literature to date has focused on VBID programs that introduce copayment reductions for
pharmaceuticals (that is, “carrot” VBID), and no
evaluation has yet been published on programs
that raise cost sharing (that is, “stick” VBID).5
Because of the growing interest in VBID and
its codification in section 2713 of the Affordable
Care Act, we updated our prior review and quantitatively synthesized data from the relatively
large number of recently published observational studies; we conclude that VBID may improve
quality of care without greatly increasing or decreasing health expenditures.5
JULY 2013
32:7
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
William H. Shrank is an
assistant professor in the
Division of
Pharmacoepidemiology and
Pharmacoeconomics, Brigham
and Women’s Hospital and
Harvard Medical School.
Niteesh K. Choudhry
([email protected])
is an associate physician
in the Division of
Pharmacoepidemiology and
Pharmacoeconomics at
Brigham and Women’s
Hospital and an associate
professor at Harvard Medical
School.
Health A ffairs
1251
Payment Models
Study Data And Methods
Data Sources We performed a structured electronic search of peer-reviewed journals using
PubMed, EconLit, Embase, Business Source
Complete, and the National Bureau of Economic
Research for studies published or in press before
November 2012 that reported on the effects of
VBID policies on medication use, medication
expenditures, and health expenditures.
Our electronic search strategy included medical subject headings (MeSH) and keywords related to pharmaceuticals (for example, “economics, pharmaceutical,” “drug utilization,” “fees,
pharmaceutical”), health care policy (for example, “health government policy regulation,”
“health policies”), and policy analysis (for example, “health economics,” “cost and cost analysis”), in addition to those specifically related
to value-based insurance design by name (for
example, “value based insurance,” “value based
benefit design,” “evidence based plan design”).
Search terms were adjusted for each database, and a common overall architecture was
maintained.
Methods We evaluated 198 abstracts to identify potentially relevant articles, including only
observational studies and excluding those that
(1) did not evaluate the effects of a value-based
policy on medications, (2) did not present original data, or (3) did not assess changes in medication adherence or expenditures—our main
outcomes of interest. We retrieved the published
version of all twenty-six candidate articles and
reviewed their reference lists to identify additional relevant studies.We contacted the primary
authors of each article to obtain additional information, including, in one case, unpublished
economic data.6
Two authors (Joy Lee, Shveta Raju) extracted
data on study populations and characteristics,
results, and study quality from each article, using a standardized protocol and reporting form,
and resolved disagreements by consensus.
Specific information collected included study
and analysis design (that is, how confounding
was controlled for), policy design (that is,
changes in copayment), patient sample (that
is, disease specific or general), drug classes, implementation date, and outcomes.
Limitations Our review had several limitations. Because of the nascent nature of the
existing literature, the review included only thirteen published studies that met our evaluation
criteria. This small sample size limited our ability to evaluate the effects of specific aspects of
value-based insurance design, such as whether
policies geared toward patients known to have a
specific disease were more effective in improving
health care quality and reducing expenditures
1252
H e a lt h A f fai r s
J U LY 201 3
3 2: 7
than policies making the benefit available to
all patients who used a particular medication.
Moreover, the review included only observational studies, which, by their nature, are somewhat limited in their internal validity. The quality of the studies and the extent to which the
studies accounted for their observational designs also differed (for example, two studies
did not have concurrent control groups). Six
evaluations were limited to one year after implementation, and eleven lacked clinical outcomes
beyond medication adherence. Finally, the
evaluated policies affected many different disease classes and drugs, and the associations differed by drug class. This heterogeneity limited
our ability to pool the data across studies to generate summary measures on the effect of valuebased insurance design.
Study Results
Our search yielded thirteen published studies
(see the online Appendix)7 that evaluated the
effects of VBID policies introduced by nine plan
sponsors, of which one was a public entity (State
of Colorado).6,8–20 In each study, copayment reductions were applied to medications used to
treat chronic diseases (Exhibit 1). Six studies
examined the effects of VBID policies on multiple drug classes. Diabetes and hypertension
medication were the most common classes
subject to copayment reductions.6,12,15,16,18,20 The
decrease in copayment ranged from 25 percent
to 100 percent per prescription (generally between $5 and $12.50).
Methodological Quality Of Observational Studies The studies were generally of good
or excellent methodological quality (see the online Appendix for more discussion).7 All but two
studies17,20 included control groups and adjusted
for covariate imbalance via covariate adjustment
in regression analysis or propensity score adjustment. The covariates available for adjustment
varied across studies to some degree, and some
studies were more explicit than others about the
variables that were adjusted for in the analysis.
Age, sex, and medication use were the most
common covariates. All studies included at least
one year of pre-period outcomes, to enable difference-in-differences analysis and strengthen
the internal validity of the quasi-experimental
evaluations.
Impact On Medication Use And Adherence
All thirteen studies reported an increase in medication adherence (an average change of 3.0 percent after one year). Copayment reductions at
one year ranged from 0.5 percent to 9.9 percent
(Exhibit 2). In two studies, however, these
changes were not statistically significant.15,18
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
Exhibit 1
Descriptions Of Value-Based Insurance Design (VBID) Policies For Prescription Drugs
Policy (year)
CVS Caremark
(2007)
Study authors
Chang et al. (Note 8
in text)
Drug class targeted
Antidiabetics
Marriott (2005)
Chernew et al.
(Notes 6 and 9 in
text)
Antidiabetics, ACE
inhibitors/ARBs,
beta-blockers,
statins, steroids
Pitney Bowes
(2007)
Choudhry et al.
(Notes 10 and 11
in text)
Choudhry et al.
(Notes 10 and 11
in text)
Gibson et al.
(Note 15 in text),
Kelly et al.
(Note 20 in text)
Novartis (2005)
Pre-VBID plan
design
3 tiers
Study patients
20,173
beneficiaries
from 3 plans
3 tiers
Eliminated for tier 1,
tier 2 reduced to
$12.50, tier 3
reduced to $22.50
37,867 employees
and dependents
Adherence
Statins
3 tiers
Eliminated for all
statins
Adherence,
cost
Clopidogrel
3 tiers
Reduced to tier 1
2,051 beneficiaries
with diabetes on
statins
779 beneficiaries
on clopidogrel
Antidiabetics,
antihypertensives,
bronchodilators
20% coinsurance for retail
scripts, 10%
coinsurance
for mail-order
scripts
10% coinsurance for
retail scripts, 7.5%
coinsurance for mailorder prescriptions
25,784 employee
beneficiaries
(Gibson et al.)
9,624 employee
beneficiaries
(Kelly et al.)
Adherence,
payment,
use
Adherence,
payment
Antidiabetics
10–35%
coinsurance
10–35%
coinsurance
10% coinsurance
1,876 employee
beneficiaries
328 employee
beneficiaries
Adherence,
payment
Adherence,
payment
747,400
beneficiaries
of participating
employers
Adherence,
cost
All drugs and testing
supplies reduced to
tier 1
Eliminated for tier 1,
increased to $35 for
tier 2 and $50 for
tier 3
589 state workers
Adherence,
utilization
4,654 beneficiaries
of groups with
50 or fewer
employees
Adherence,
cost
Most items reduced to
tier 1
Reduced copayments
by 50% for 3
preferred drugs
71 beneficiaries
of one employera
14,976
beneficiaries
from 24
employersponsored plans
Adherence
Florida Health
Care Coalition
(2006)
Gibson et al.
(Note 14 in text)
Blue Cross Blue
Shield of
North
Carolina
(2008)
Maciejewski et al.
(Note 16 in text),
Farley et al.
(Note 12
in text)
Antidiabetics,
antihypertensives,
cholesterollowering
medications
3 tiers
State of
Colorado
(2006)
Blue Cross Blue
Shield of
Minnesota
(2006)
Nair et al. (Note 17
in text)
Antidiabetics
3 tiers
Rodin et al. (Note 18
in text)
Antidiabetics,
cardiac
medications
3 tiers
Zeng et al. (Note 19
in text)
Frank et al. (Note 13
in text)
Antidiabetics
3 tiers
Statins
3 tiers
Health Alliance
(2007)
Health Alliance
(2008)
Outcomes
evaluated
Adherence
Copay description
Copay reductions for
tier 1 and tier 2
Antidiabetics
10% coinsurance with
disease
management
Eliminated for tier 1
for program
participants,
reduced for tiers 2
and 3 for all
beneficiaries
Adherence,
cost
Adherence
SOURCE Authors’ analysis of cited studies. NOTES ACE is angiotensin-converting enzyme. ARB is angiotensin receptor blocker. aCarle Clinic, part of Health Alliance.
Five studies that examined multiple drug classes
found variations in adherence response across
drug classes within the same studies.9,12,15,17,20 The
largest adherence improvements were found for
diabetes medications, followed by statins, angiotensin-converting enzyme (ACE) inhibitors, and
angiotensin receptor blockers (ARBs).14,19,21
Four studies also reported adherence changes
beyond one year (Exhibit 3). Teresa Gibson and
colleagues found that for diabetes medications,
adherence decreased slightly in the first year
after program implementation (−0.2 percent)
but improved by year 3.15 Kavita Nair and colleagues found that compared to the year prior
to copayment elimination, adherence improved
by 9.4 percent in the first year after copayments
were eliminated for insulin and by 11.3 percent a
year later.17 Gibson and colleagues’ evaluation of
the Florida Health Care Coalition’s VBID program found that for patients in a disease manJULY 2013
32:7
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
Health A ffairs
1253
Payment Models
Exhibit 2
Impact Of Value-Based Insurance Design On Patients’ Adherence With Prescribed
Medications
Change in adherence (%)
Policy
CVS/Caremark
Drug class
Insulin
Oral antidiabetics
All antidiabetics
At 1 year
9.9f
5.0f
7.2f
At 3 years
—
—
—
Marriott
ACE inhibitors, ARBs
Beta-blockers
Antidiabetics
Statins
Inhaled steroids
2.6f
3.0f
4.0f
3.4f
1.9a
—
—
—
—
—
Pitney Bowes
Statins
Clopidogrel
3.1d
4.2d
—
—
Novartis
Asthma
Antidiabetics
Cardiovascular
Overall
Asthma
Antihypertensives
Antidiabetics
Florida Health
Care Coalition
Insulin
Oral antidiabetics
All diabetes medications
Insulin with disease
management
Oral antidiabetics with
disease management
All diabetes medications
with disease management
0.0b
−0.2e
0.7e
0.5e
—
—
—
0.3b
2.0d
−0.1e
1.8e
9.0c
9.0c
4.0c
2.5b
0.4b
8.3b
3.6b
0.8b
4.1b
3.7b
2.7d
1.0e
5.8f
3.7e
6.5f
ACE inhibitors, ARBs
ARBs
Beta-blockers
Calcium channel blockers
Cholesterol absorption
inhibitors
Diuretics
Metformin
Statins
2.5
0.9e
2.2f
0.9e
4.8a,f
−0.2a,b
4.3a,f
2.0a,f
0.3b
2.8f
3.2f
1.4f
0.4a,b
4.5a,f
5.0a,f
2.3a,f
State of Colorado
Insulin
Oral antidiabetics
Overall
9.4e
2.5c
3.2c
11.3a,e
−1.0c
1.3c
Blue Cross Blue Shield
of Minnesota
Health Alliance 2007
Statins
Sulfonylurea
Metformin
Thiazolidinediones
Insulin
Antidiabetics
4.9d
0.6c
2.3c
−1.9c
−0.6c
7.3f
—
—
—
—
—
—
Health Alliance 2008
Statins
2.7d
—
Blue Cross Blue Shield
of North Carolina
f
SOURCE Authors’ analysis of cited studies (see Exhibit 1). NOTES In empty cells, the study did not
assess an outcome for that time frame. ACE is angiotensin-converting enzyme. ARB is angiotensin
receptor blocker. aAt two years. bNot significant. cp value not available. dp < 0:05 ep < 0:01
f
p < 0:001
agement program, the effect increased from
3.7 percent in the first year to 6.5 percent by
the third year.14 In contrast, patients in the
VBID program implemented without disease
management saw nonsignificant changes in all
three years. The study by Joel Farley and col1254
H e a lt h A f fai r s
J U LY 2 0 1 3
3 2: 7
leagues found that adherence increased 3–
9.7 percent (depending on medication) for previously nonadherent patients between the first
and second years of the VBID program.12
Impact On Prescription And Health
Spending Several studies evaluated the effect
of VBID policies on health plan spending. As
expected, VBID policies were associated with significant increases in drug spending for insurers
in the five studies that examined this outcome
(Exhibit 3). This ranged from a 0.2 percent increase in expenditures for diabetes, hypertension, and asthma medications at Novartis15 to a
61 percent increase in expenditures for diabetes
medications for the State of Colorado.17
In the four studies that reported nonprescription medical spending for insurers, the VBID
policies were not associated with significant
changes in health spending.9,11,14,15 Two studies
that bundled VBID and disease management
found that expenditures trended lower.6,14 In
one, Michael Chernew and colleagues reported
medical savings of $51.03 per member per
month, although the statistical significance of
this change was not reported.6 In the other,
Gibson and colleagues also reported statistically
insignificant savings.14 In that study, only patients with diabetes who enrolled in disease management programs saw a significant drop in
medical expenditures (26.4 percent reduction,
p < 0:10). The trend held at three years (40.5 percent reduction, p < 0:05).14
Six studies evaluated the effect of the policies
on medical and prescription spending collectively, and none observed significant increases
past the first year.6,11,14,15,17,20 Chernew and colleagues, Choudhry and colleagues, Gibson and
colleagues, and Emily Kelly and colleagues did
not observe statistically significant changes in
overall insurer expenditures,9,11,15,20 which suggests that the VBID policies may have increased
prescription spending without increasing overall spending. Two studies observed significant
changes in the first year, but not in the years
after. Nair and colleagues observed an 18 percent
increase in expenditures after one year
(p < 0:05) compared to baseline. In the following year, the expenditures decreased by 18 percent compared to baseline (p < 0:05).17 In contrast, Gibson and colleagues saw a 20.3 percent
decrease in overall expenditures in a disease
management setting after one year (p < 0:01).
At three years the spending decrease shrank to
11.8 percent and was no longer statistically
significant.14
Impact On Health Services Use Two of the
studies examined the impact of value-based insurance design on health services use. Nair and
colleagues found that the policy was associated
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
Exhibit 3
Impact Of Value-Based Insurance Design On Prescription And Medical Expenditures
Change in expenditures
Policy
Study author
Drug class
At 1 year
At 3 years
Prescription expenditures
Marriotta
Chernew et al.
Antidiabetics, ACE inhibitors/ARBs,
beta-blockers, statins, and steroids
Statins
Clopidogrel
Antidiabetics, asthma, and antihypertensives
Asthma
Antihypertensives
Antidiabetics
Diabetes
All causes
Diabetes with disease management
All causes with disease management
Not specified
$20.87 PMPMb
Pitney Bowes
Choudhry et al.
Novartis
Gibson et al.
Kelly et al.
Florida Health
Care Coalition
Gibson et al.
State of Colorado
Nair et al.
Antidiabetics, ACE inhibitors/ARBs,
beta-blockers, statins, and steroids
Pitney Bowes
Choudhry et al.
Statins
Clopidogrel
Novartis
Gibson et al.
Antidiabetics, asthma, and antihypertensives
Florida Health
Gibson et al.
Diabetes
Care Coalition
All causes
Diabetes with disease management
All causes with disease management
Prescription and medical expenditures
−$51.03 PMPMb
Marriotta
Chernew et al.
−$26.88 PMPMb
Pitney Bowes
Choudhry et al.
Novartis
Gibson et al.
Kelly et al.
Florida Health
Care Coalition
Gibson et al.
State of Colorado
Nair et al.
0.0%c
−6.0c
0.2c
2.5b
19.9b
10.5b
−5.8c
8.3c
15.7e
12.5e
61.0d
—
—
—
16.6%c
—
—
—
−16.0c
23.0c
17.7d
21.6e
—
Medical expenditures
Marriotta
Chernew et al.
Antidiabetics, ACE inhibitors/ARBs,
beta-blockers, statins, and steroids
Statins
Clopidogrel
Antidiabetics, asthma, and antihypertensives
Asthma
Antihypertensives
Antidiabetics
Diabetes
All causes
Diabetes with disease management
All causes with disease management
Antidiabetics
16.0%c
−26.0b
34.0c
−4.3c
−30.6f
−26.4c
−1.9c
3.0%c
−6.0c
28.0c
—
—
—
−4.0c
−20.3e
−6.6c
3.3c
18.0d
—
—
—
8.9%c
25.6c
−23.5c
−40.5d
−0.2c
—
—
—
8.4%c
−18.0b
−5.0b
−9.0b
3.4c
−11.8c
−15.3c
8.5c
—
SOURCE Authors’ analysis of cited studies (see Exhibit 1). NOTES In empty cells, the study did not assess an outcome for that time
frame. ACE is angiotensin-converting enzyme. ARB is angiotensin receptor blocker. PMPM is per member per month. aBased on
unpublished data. bp value not available. cNot statistically significant. dp < 0:05 ep < 0:01 fp < 0:001
with significant decreases in emergency department visits (−36 percent, p < 0:01), physician
office visits (−5 percent), and hospitalizations
(−13 percent) at two years.17 Likewise, Choudhry
and colleagues observed significant reductions
in rates of physician visits for statin and clopidogrel users after copayment reduction (relative
change: statin users: 0.80, 95% confidence interval: 0.57, 0.98; clopidogrel users: 0.87, 95%
CI: 0.59, 0.96). Reductions in hospitalizations
and emergency department admissions were
also observed (relative change: statin users:
0.90, 95% CI: 0.80, 0.92; clopidogrel users:
0.89, 95% CI: 0.74, 0.90).11
Discussion
This review of observational evaluations found
that value-based insurance design was consistently associated with improved medication adherence. However, consistent with the limited
randomized trial evidence evaluating valuebased insurance design,22 our review also found
that the improvements in medication adherence
associated with it were not accompanied by
significant reductions in overall medical or total
insurer spending.
The primary benefit of value-based insurance
design may be in its ability to improve the quality
of care for patients with chronic diseases.
JULY 2013
32:7
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
Health A ffairs
1255
Payment Models
However, much of the enthusiasm for this approach has been based on the hope that payer
spending would decline after implementation.23
This expectation comes from previous research
based upon simulations from economic models
that examined lifetime trends (not one-to-threeyear trends as in these studies).23 For example,
using Medicare data, Choudhry and colleagues
estimated the cost-effectiveness of eliminating
all cost sharing for prescription drugs to treat
Medicare beneficiaries following myocardial infarction. They found that this strategy would be
cost saving within one year after the initial myocardial infarction and would lead to total savings of $2,500 per beneficiary over beneficiaries’
remaining life span.19 Two other studies reached
similar conclusions regarding full coverage of
ACE inhibitors for Medicare patients with diabetes.23,24 Yet in our review we did not find that
these cost savings were realized in real-world
VBID implementations in the short term (one
to three years).6,11,14,15,17,20
VBID programs can employ a variety of methods to reward the use of therapies of high clinical
value, from reducing copayments and coinsurance rates to shifting medications to a lower tier
(to lower beneficiary cost sharing) in the drug
formulary. We found no published examples of
VBID policies that discourage the use of lowvalue therapies. However, policies such as
higher cost sharing for certain cancer medications that are administered outside of guideline
recommendations have been proposed.25 Determining what services or treatments are defined
as low value may be a more difficult and controversial process than that required to define high
value services. Yet both approaches aim to deter
the use of treatments with uncertain or limited
effectiveness and to encourage evidence-based
practices.25
Many insurers recognize that VBID programs
are a part of a larger strategy to promote the use
of services of high clinical value. Consequently,
some VBID programs link copayment reductions
to clinical criteria or forms of patient engagement. For example, Gibson and colleagues found
greater improvements in adherence from VBID
programs only among patients who were enrolled in a disease management program.14
Given the wide variety of choices and the paucity
1256
Health Affairs
JULY 2013
32:7
of existing data evaluating programs other than
copayment reductions, the optimal design of
VBID programs to obtain the greatest positive
effects remains to be determined.5
Many unanswered questions remain regarding how to measure and maximize the potential
benefits of VBID programs. The existing literature provides limited information about whether
value-based insurance design improves health
outcomes. The relatively short duration of many
studies may have been insufficient to capture
health effects that may take years to become apparent. The impact of value-based insurance design may also differ by the disease and risk levels
of patients who are targeted. Moreover, the
existing evaluation literature focuses solely on
health spending, which provides an important
but incomplete assessment of the business case
for value-based insurance design. Employers
considering VBID implementation may be interested in additional impacts, such as the impact
on employee productivity (either missing work
or suffering lower productivity while at work
because of illness).
Conclusion
Implementation of value-based insurance design
has been recommended as a possible contributor
to efforts to improve health care quality without
increasing cost and is of great political and research interest.25 The Affordable Care Act included a provision that “the Secretary may develop guidelines to permit a group health plan
and a health insurance issuer offering group or
individual health insurance coverage to utilize
value-based insurance designs.”26 Our review
suggests that although value-based insurance design might not significantly reduce health spending in the short term—that is, within one to three
years—some VBID plans improve medication adherence and reduce patients’ out-of-pocket expenses. Clearly, studies that examine the longer-term, real-life effects of value-based insurance
design are much needed to see if adherence remains improved and assess its impact on overall
insurer expenditures. Efforts to identify the optimal VBID design should be the focus of ongoing research to realize the potential of this
approach to quality improvement. ▪
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
The authors received research support
through an unrestricted grant to
Brigham and Women’s Hospital from
Aetna Inc. for a trial evaluating the
impact of cost-sharing reductions on
cardiovascular outcomes, and from CVS
Caremark to study medication
adherence. Niteesh Choudhry is a
consultant to Mercer Health and
Benefits Inc. Matthew Maciejewski was
supported by the Robert Wood Johnson
Foundation’s Changes in Health Care
Financing and Organization Initiative
(Grant No. 67461), Blue Cross Blue
Shield of North Carolina, and a Research
Career Scientist Award from the
Department of Veterans Affairs
(RCS 10-391). Maciejewski has received
consultation funds from Daichi Sankyo,
Takeda Pharmaceuticals, Novartis,
ResDAC at the University of Minnesota,
and the Surgical Review Corporation,
and he owns stock in Amgen through his
spouse’s employment.
Intern Med. 2010;25:S381–2.
11 Choudhry NK, Fischer MA, Avorn JL,
Lee JL, Schneeweiss S, Solomon DH,
et al. The impact of reducing
cardiovascular medication copayments on health spending and resource utilization. J Am Coll Cardiol.
2012;60(18):1817–24.
12 Farley JF, Wansink D, Lindquist JH,
Parker JC, Maciejewski ML.
Medication adherence changes following value-based insurance design. Am J Manag Care. 2012;18(5):
265–74.
13 Frank MB, Fendrick AM, He Y,
Zbrozek A, Holtz N, Leung S, et al.
The effect of a large regional health
plan’s value-based insurance design
program on statin use. Med Care.
2012;50(11):934–9.
14 Gibson TB, Mahoney J, Ranghell K,
Cherney BJ, McElwee N. Value-based
insurance plus disease management
increased medication use and produced savings. Health Aff
(Millwood). 2011;30(1):100–8.
15 Gibson TB, Wang S, Kelly E, Brown
C, Turner C, Frech-Tamas F, et al. A
value-based insurance design program at a large company boosted
medication adherence for employees
with chronic illnesses. Health Aff
(Millwood). 2011;30(1):109–17.
16 Maciejewski ML, Farley JF, Parker J,
Wansink D. Copayment reductions
generate greater medication adherence in targeted patients. Health Aff
(Millwood). 2010;29(11):2002–8.
17 Nair KV, Miller K, Park J, Allen RR,
Saseen JJ, Biddle V. Prescription copay reduction program for diabetic
employees. Popul Health Manag.
2010;13(5):235–45.
18 Rodin HA, Heaton AH, Wilson AR,
Garrett NA, Plocher DW. Plan designs that encourage the use of
generic drugs over brand-name
drugs: an analysis of a free generic
benefit. Am J Manag Care. 2009;
15(12):881–8.
19 Zeng F, An JJ, Scully R, Barrington
C, Patel BV, Nichol MB. The impact
of value-based benefit design on
adherence to diabetes medications: a
propensity score-weighted difference in difference evaluation. Value
Health. 2010;13(6):846–52.
20 Kelly E, Turner C, Frech-Tamas F,
Doyle J, Mauceri E. Value-based
benefit design and healthcare utilization in asthma, hypertension,
and diabetes. Am J Pharm Benefits.
2009;1(4):217–21.
21 Chernew ME, Shah MR, Wegh A,
Rosenberg SN, Juster IA, Rosen AB,
et al. Impact of decreasing copayments on medication adherence
within a disease management environment. Health Aff (Millwood).
2008;27(1):103–12.
22 Choudhry NK, Avorn J, Glynn RJ,
Antman EM, Schneeweiss S,
Toscano M, et al. Full coverage for
preventive medications after myocardial infarction. N Engl J Med.
2011;365(22):2088–97.
23 Choudhry NK, Patrick AR, Antman
EM, Avorn J, Shrank WH. Cost-effectiveness of providing full drug
coverage to increase medication adherence in post-myocardial infarction Medicare beneficiaries.
Circulation. 2008;117(10):1261–8.
24 Rosen AB, Hamel MB, Weinstein
MC, Cutler DM, Fendrick AM, Vijan
S. Cost-effectiveness of full Medicare
coverage of angiotensin-converting
enzyme inhibitors for beneficiaries
with diabetes. Ann Intern Med.
2005;143(2):89–99.
25 Neumann PJ, Auerbach HR, Cohen
JT, Greenberg D. Low-value services
in value-based insurance design. Am
J Manag Care. 2010;16(4):280–6.
26 Fairman KA, Curtiss FR. What do we
really know about VBID? Quality of
the evidence and ethical considerations for health plan sponsors. J
Manag Care Pharm. 2011;17(2):
156–74.
NOTES
1 Shrank WH, Asch SM, Adams J,
Setodji C, Kerr EA, Keesey J, et al.
The quality of pharmacologic care
for adults in the United States. Med
Care. 2006;44(10):936–45.
2 Choudhry NK. Copayment levels and
medication adherence: less is more.
Circulation. 2009;119(3):365–7.
3 Brennan T, Reisman L. Value-based
insurance design and the next generation of consumer-driven health
care. Health Aff (Millwood). 2007;
26(2):w204–7. DOI: 10.1377/
hlthaff.26.2.w204.
4 Cranor CW, Bunting BA, Christensen
DB. The Asheville Project: long-term
clinical and economic outcomes of a
community pharmacy diabetes care
program. J Am Pharm Assoc (Wash).
2003;43(2):173–84.
5 Choudhry NK, Rosenthal MB,
Milstein A. Assessing the evidence
for value-based insurance design.
Health Aff (Millwood). 2010;29(11):
1988–94.
6 Chernew ME, Juster IA, Shah M,
Wegh A, Rosenberg S, Rosen A, et al.
The financial effects of a value based
insurance design program. 2009.
Unpublished paper.
7 To access the Appendix, click on the
Appendix link in the box to the right
of the article online.
8 Chang A, Liberman JN, Coulen C,
Berger JE, Brennan TA. Value-based
insurance design and antidiabetic
medication adherence. Am J Pharm
Benefits. 2010;2(1):39–44.
9 Chernew ME, Shah MR, Wegh A,
Rosenberg SN, Juster IA, Rosen AB,
et al. Impact of decreasing copayments on medication adherence
within a disease management environment. Health Aff (Millwood).
2008;27(1):103–12.
10 Choudhry N, Fischer M, Liu J, Lii J,
Brookhart A, Shrank W, et al.
Reducing copayments for essential
cardiovascular medications: evaluation of a natural experiment. J Gen
J U LY 2 0 1 3
32:7
Downloaded from content.healthaffairs.org by Health Affairs on July 9, 2013
at WELCH MEDICAL LIBRARY JHU
Health Affairs
1257